高分辨率可见光遥感图像舰船目标检测综述

宋志娜; 眭海刚; 李永成

doi:10.13203/j.whugis20200481

高分辨率可见光遥感图像舰船目标检测综述

1.
武汉大学遥感信息工程学院，湖北武汉，430079
2.
武汉大学测绘遥感信息工程国家重点实验室，湖北武汉，430079

基金项目:

国家自然科学基金 41771457

国家重点研发计划 2018YFB10046

湖北省自然资源厅科技项目 ZRZY2020KJ03

详细信息

作者简介:
宋志娜，博士，主要从事遥感图像目标信息提取研究。song.zhina@163.com

通讯作者:
眭海刚，博士，教授。haigang_sui@263.net

中图分类号: P237
计量
- 文章访问数: 03284
- HTML全文浏览量: 0863
- PDF下载量: 0495
出版历程
- 收稿日期: 2020-09-10
- 发布日期: 2021-11-04

A Survey on Ship Detection Technology in High-Resolution Optical Remote Sensing Images

1.
School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China
2.
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China

Funds:

The National Natural Science Foundation of China 41771457

the National Key Research and Development Program of China 2018YFB10046

Research Program of the Department of the Natural Resources of Hubei Province of China ZRZY2020KJ03

More Information

Author Bio:
SONG Zhina, PhD, specializes in extracting target information from remote sensing images. E-mail: song.zhina@163.com

Corresponding author:
SUI Haigang, PhD, professor. E-mail: haigang_sui@263.net

摘要

摘要: 利用大规模、高分辨率可见光遥感图像探测舰船目标一直是遥感领域的研究热点，其在军事和民用领域发挥着重要作用。舰船目标背景的复杂多变性，舰船目标的多尺度、多种类、多姿态类内差异性，以及检测模型的局限性等问题，都给大范围的高分辨率可见光遥感图像舰船目标探测带来了极大的挑战。首先对利用遥感技术进行舰船检测的方法体系进行总结归纳，然后在此基础上聚焦当前研究热点，对2010—2020年高分可见光遥感图像舰船目标的检测技术体系与发展历程进行概述，并对主流的检测方法进行详细论述，以期推动高分辨率可见光遥感影像舰船检测更加深入的研究和更加广泛的应用。
- 高分辨率影像 /
- 可见光遥感影像 /
- 舰船目标检测 /
- 自动目标解译
Abstract:
Objectives Ship detection in large-scale, high-resolution optical remote sensing images, especially in visible spectral remote sensing images, plays an important role in both military and civilian fields. It has always been an on-going and challenging research topic in the past decades, due to the complexity and variability of the backgrounds, the multi-scale, multi-type, and multi-posture diversity of ship targets.
Methods In order to promote the development of ship detection based on high-resolution optical remote sensing images (HRORSI), this paper provides an overview of existing study on ship detection methods of HRORSI over the past decades. Firstly, we introduce all the existing monitoring systems for ship detection, but subsequently focus only on HRORSI. Then, a number of topics have been discussed, including the methodology system, the development history, the recent state of the art detection methods, detection datasets and metrics.
Results It is shown that current ship detection methods based on deep learning in HRORSI, have greatly expanded the adaptability of the complexity of the detection scene and the variation of the target distribution, the speed and accuracy of the results have been greatly improved.It is also found that different influence factors make a big difference in choosing the corresponding ship detection models.
Conclusions Existing methods from different perspectives such as extracting richer features, multi-level and multi-scale feature fusion, more accurate target locating, scale aware ship detection, have made vigorous performance.However, there is still a big gap between efficient and intelligent ship interpretation in actual complex applications.We suggest that the future ship detection methods should adaptively support a variety of backgrounds, scales and optical sensors, the detection model can be effectively transfer to out-distributed target domain scenarios as well as resource-constrained scenarios, and more in-depth target recognition capabilities.
- high-resolution remote sensing images /
- optical remote sensing image /
- ship detection /
- automatic target interpretation

HTML全文

遥感变化检测是利用不同时相的遥感影像和相关地理数据，结合遥感成像机理和地物特性，采用图像、图形处理理论和数学模型方法，确定和分析研究区的地表覆盖变化范围和变化类型。其研究目的是提取研究区变化信息，并生成变化图^[1-2]。

高分辨率遥感影像比中低分辨率遥感影像具备更加丰富的光谱、纹理和形状等地物特征，但影像中同类地物的差异性增强，不同地物的光谱特征相互混淆，影像光谱域的统计可分性降低，信息提取难度增大，影像中“同物异谱，异物同谱”的现象大量发生，加重了地物光谱和纹理特征的离散程度，使得基于像元的变化检测精度难以提高^[3]。

深度学习^[4-5]作为机器学习的一个重要分支，可以自动地将简单特征组合成复杂特征，并利用复杂特征进行分类，分类精度大大提高，其在变化检测领域的应用比其他方法更为宽泛，适用性也更强。随着AlexNet^[6]、VGG^[7]、GoogLeNet^[8]和ResNet^[9-10]等深度学习网络的出现，神经网络在变化检测领域得到了成功应用。张鑫龙等^[2]提出了基于深度玻尔兹曼机的深度学习变化检测方法；Peng等^[10]提出了一种改进的用于语义分割编码器-解码器结构，端到端变化检测方法；Mou等^[11]将光谱空间模块和时态模块进行组合，设计了一种循环3D全卷积网络用于建筑物变化检测。虽然国内外学者针对变化检测中存在的问题进行了大量研究，但深度学习变化检测方法优质训练样本的选取和自动化程度、基于像元检测的“椒盐”现象和伪变化区的出现等仍是目前亟待解决的问题^[12-14]。

领域知识在遥感影像变化检测中的应用，对于解决图像处理算法的局限性、提高变化检测的精度和图像处理的自动化程度具有较为理想的效果^[3]。深度置信网络（deep belief networks，DBN）是深度学习领域较为经典的模型，它通过联合概率分布来推断样本数据分布，非常适用于基于像元的变化检测研究。改进的变化矢量分析算法^[2]（robust change vector analysis，RCVA）和灰度共生矩阵算法^[15-16]（grey level co-occurrence matrix，GLCM）能够提取影像光谱和纹理差异特征，减小由于预处理阶段的配准所造成的误差，减弱高分辨率遥感影像中地物间光谱特征相互混淆等不利因素对检测结果的影响。

本文以DBN为深度学习模型，利用RCVA和GLCM算法提取影像光谱和纹理差异特征。将原理性和经验性的领域知识整合成为知识规则，用于选取优质训练样本并作为优化深度学习变化检测结果的依据。通过高分二号与IKONOS影像的变化检测实验，验证了本文方法的有效性。

1 本文方法

1.1 光谱变化特征提取

由于不同时相影像成像条件不同，在对影像进行精配准后，影像间的配准误差仍难以消除。其结果则是两幅影像间的像元对应关系不正确，进而导致了检测时伪变化区的大量出现。RCVA算法通过考虑像元的邻域信息，选择光谱差异最小的像元对进行检测，消除了配准误差带来的影响。

RCVA算法原理如下：基于影像1中的某个像元 $x_{1} (j, k)$ ，在影像2 $(j \pm ω, k \pm ω)$ 范围内求与 $x_{1}$ 亮度值差异最小的像元 $x_{2}$ ，此时认为 $x_{2}$ 为 $x_{1}$ 的同名像点，并求取差值 $x_{d i f f e r e n t_a}$ ，表示通过亮度值求得的影像1中 $(j, k)$ 点的变化强度值。同样，基于影像2求对应影像1中的同名像点，并求取差值 $x_{d i f f e r e n t_b}$ ，表示通过亮度值求得的影像2中 $(j, k)$ 点的变化强度值，以 $x_{d i f f e r e n t_a}$ 和 $x_{d i f f e r e n t_b}$ 较小者作为该点变化强度。

遍历影像，可得到所有像元的光谱变化强度值，进而得到考虑邻域信息的光谱变化强度图。

1.2 纹理变化特征提取

纹理特征是反映图像中同质的一种视觉特征，它体现了物体表面的结构组织排列属性，对于反映物体的表层特征变化具有重要利用价值。GLCM是提取纹理的一种经典方法，也是目前普遍使用且提取效果较好的纹理特征分析法^[16]。已有的研究定义了14种标量来进行纹理分析，其中最常用的有均值、方差、协同性、对比度、熵等8 ‍种。以方差为标量研究纹理特征时，最能反应不同地物间的差异^[17-18]。

得到两幅影像的方差特征值后，即可通过差值计算得到纹理变化强度图。

1.3 结合领域知识的训练样本优化

根据光谱变化和纹理变化强度图，通过设置不同阈值提取样本，可对样本进行不同程度的划分。为保证样本的充分性，最大限度地选择到具有代表性的变化地物和未变化地物样本，通过自定义阈值取并集的方式，分析得出合理的阈值组合，并选择该阈值组合下的标记样本作为待优化样本。

阈值组合的选取如下：以光谱变化和纹理变化强度图为基准，在最小强度值和最大强度值闭区间内，从小到大逐个设置阈值对样本进行标记；以参考变化结果为依据，对所有阈值下的样本标记结果进行统计，计算其准确率；计算相邻阈值间的准确率增幅，由于准确率反映了正负样本标记的正确程度，因此在准确率增幅趋于稳定前的节点处即可选定较为合理的阈值；以相同的方法分别获得RCVA和GLCM下的合理阈值，即组成阈值组合。通过变化强度区间逐个分析得到合理阈值的方式，对阈值的分析较细致，同样能够运用于其他情景下的阈值分析。

高分辨率遥感影像中异常光谱值多，同时，由于影像获取时间和获取条件的不同，植被季节性返青和建筑物阴影区等因素影响，造成了伪变化样本的大量出现。

1）形状特征知识。本文结合领域知识，引入面积（S）和形状复杂度（C）两种形状特征指数作为过滤离散伪变化样本的优化策略。优化策略定义为：（1） $S \geq S_{m i n}$ ，即所有变化样本组成的图斑，其面积均应大于等于最小定义图斑面积，若不满足，则将其归并为未变化样本；（2） $(C = P / s) \leq C_{m i n}$ ， $P$ 为待检测图斑周长， $s$ 为其面积，当满足条件 $S_{m a x} \geq s \geq S_{m i n}$ 时， $C$ 值越大说明图斑越复杂，是离散碎图斑的概率就越大，可将其归并为未变化样本。根据形状特征知识，本文设计的各类小图斑检测图谱如图 1所示。

图 1 小图斑检测图谱

Figure 1. Graph for Detecting Small Patches

下载: 全尺寸图片幻灯片

图 1中共有134种图斑类型，是像素个数不大于4的所有图斑形式，其中，（a）~（l）为根据图斑形状和遍历窗口大小进行的排序。将待检测图斑与图 1比对，若形状相同，则可对该图斑进行删除或归并（归为未变化图斑）。

2）光谱特征知识。针对植被（农作物、绿化带等）季节性返青和建筑物阴影等因素所引起的带状、块状伪变化样本集中区，通过对两幅影像固定窗口内同名像元亮度值采样的方式，自定义光谱映射区间。依据光谱映射区间建立光谱知识规则，并对两幅影像进行遍历，剔除伪变化样本。

映射关系建立方式为：

(φ_{m i n}^{1} \leq φ^{1} \leq φ_{m a x}^{1}) \overset{f_{ω}}{\leftrightarrow} (φ_{m i n}^{2} \leq φ^{2} \leq φ_{m a x}^{2})

式中， $φ^{1}$ 和 $φ^{2}$ 分别为两幅影像中同名像元红、绿、蓝3个波段的亮度值 $(φ_{R}^{1}, φ_{G}^{1}, φ_{B}^{1} 与 φ_{R}^{2}, φ_{G}^{2}, φ_{B}^{2})$ ； $φ_{m i n}^{1}$ 与 $φ_{m i n}^{2}$ 为映射区间下限； $φ_{m a x}^{1}$ 与 $φ_{m a x}^{2}$ 为映射区间上限； $ω$ 为滑动窗口大小。

1.4 DBN模型训练与变化检测

DBN是一种概率生成模型^[19]，通过联合概率分布来推断样本数据分布。通过逐层非监督的训练方式对大量无标签样本数据进行特征提取，并通过少量有标签的样本数据进行模型优化，最后获得网络最优权重，使得网络能依据最大概率生成训练数据。DBN主要由两部分构成，第一是多层限制玻尔兹曼机（restricted Boltzmann machine，RBM），用于预训练网络；第二是前馈反向传播网络，此部分可以使RBM堆叠的网络更加精细化。

RBM含有两层（显层 $v$ 、隐层 $h$ ），为无向图模型，每层可定义为一个向量，向量的维度即为该层神经元的个数，不同层之间的神经元由权值矩阵 $W$ 连接。对于每一个RBM，都有其作为一个系统所具备的能量，而根据能量函数则可以得到关于 $v$ 与 $h$ 的联合概率分布 $P (v, h | θ)$ （ $θ = \{W_{i j}, a_{i}, b_{j}\}$ 为模型参数， $W_{i j}$ 为神经元 $i$ 、 $j$ 之间的权重， $a_{i}$ 和 $b_{i}$ 分别表示显层和隐层神经元间的偏置），DBN则是通过基于 $P (v, h | θ)$ 的相关计算来重构样本数据。经过迭代可不断优化网络参数，达到初始化每个RBM参数的目的。当RBM网络中的特征向量在映射到不同特征空间时能够保存最多的特征信息，完成DBN的预训练。基于误差反向传播算法，利用有标签的样本数据，通过计算各RBM层的学习误差，对网络权值 $W$ 进行更新，微调整个模型，完成DBN的训练。

RBM的显层用来接收变量，且传统的输入多为二值变量，而连续输入的实值变量则更适用于图像分类和语音识别等^[2]。本文DBN的输入为2×2像素范围内像元红、绿、蓝3波段的归一化亮度值依次排列的向量。

利用预选好的样本对模型进行训练，当训练精度达到精度要求后，对模型进行保存，并利用模型对整幅影像进行检测，得到变化检测结果。由于DBN模型是基于像元的变化检测，模型训练完毕后权值已固化，因此当进行变化检测时会出现少量的“椒盐”噪点。为进一步提高检测精度，利用图 1中的（a）检测图斑对未变化区域中的孤立点进行剔除，对变化区域中的孤立点进行填充，得到最终的变化检测结果。

2 实验与分析

2.1 实验1

实验1所用数据为2016-05-19和2017-04-29获取的长春市某地区高分二号遥感影像，影像分辨率为1 m，包含红、绿、蓝3个波段，影像大小为1 ‍389×2 200像素，如图 2（a）所示；利用RCVA和GLCM得到变化强度图如图 2（b）所示。通过分析准确率增幅，选定RCVA和GLCM的阈值分别为75和16，作为划分变化样本的尺度。准确率增幅随阈值变化趋势图如图 3所示。在该阈值组合下得到的变化样本，覆盖范围较广，各类变化地物样本标记较全面、充分，变化样本标记情况如图 4（a）所示。

图 2 高分二号遥感影像与变化强度图

Figure 2. GF-2 Images and Intensity Change Images

下载: 全尺寸图片幻灯片

图 3 准确率增幅变化趋势图

Figure 3. Trend of Accuracy Increase

下载: 全尺寸图片幻灯片

图 4 执行优化策略前后变化样本标记图(实验1)

Figure 4. Change Sample Marker Image Before and After Executing the Optimization Strategy (Experiment 1)

下载: 全尺寸图片幻灯片

利用领域知识优化策略，对所标记的变化样本进行过滤，得到能够准确反映变化与未变化区域的高质量样本。对植被季节性返青区和建筑物阴影区的典型区域进行采样，建立映射区间，剔除伪变化区。执行优化策略后的变化样本标记图如图 4（b）所示，其中，红色实线框内为建筑物阴影被大量剔除的典型区域，红色虚线框内为植被季节性返青大量剔除的典型区域。与图‍ 4‍（a）对比可知，执行优化策略后有大量伪变化样本被剔除。

根据图 4（b）的样本标记图，共选取变化样本211 637个，未变化样本2 844 163个。为了验证本文方法的有效性，分别在已有的变化与未变化样本中随机选取了不同数量的样本对DBN进行训练，并统计了DBN变化检测结果的准确率、召回率、虚警率和漏检率，如表 1所示，其中，A、B、C分别代表训练样本优化前、训练样本优化后和变化检测结果优化后。

表 1 训练样本优化前后精度分析表(实验1)

Table 1. Precision Analysis Before and After Optimizing Training Samples (Experiment 1)

数量/个		准确率/%			召回率/%			虚警率/%			漏检率/%
正样本	负样本	A	B	C	A	B	C	A	B	C	A	B	C
5 000	5 000	88.70	93.28	93.77	65.30	76.30	80.10	60.42	42.42	39.76	34.70	23.70	19.90
10 000	10 000	88.63	94.09	94.59	66.59	73.75	78.68	60.45	37.37	34.30	33.41	26.25	21.32
30 000	30 000	88.26	92.45	93.07	66.60	78.95	83.68	61.46	46.57	43.70	33.40	21.05	16.32
50 000	50 000	87.72	91.74	92.80	68.21	79.16	83.45	62.62	49.44	44.96	31.79	20.84	16.55
80 000	80 000	86.90	94.42	94.48	69.14	73.14	75.21	64.37	35.09	34.15	30.86	26.86	24.79

下载: 导出CSV

| 显示表格

由表 1可知，训练样本优化后，准确率和召回率有较大幅度提高，其中，准确率最大增幅7.52%，召回率最大增幅12.35%；虚警率和漏检率有较大幅度下降，其中，虚警率最大降幅29.28%，漏检率最大降幅12.35%。对变化检测结果优化后，准确率最大增幅1.06%，召回率最大增幅4.93%，虚警率最大降幅4.48%，漏检率最大降幅4.93%。

参考变化以及变化检测结果如图 5所示。其中，参考变化结果为遥感图像解译人员通过目视解译获得，并在实地进行了调绘。通过实地调绘结果的对比与补充，参考变化结果与实地变化情况符合度较高（精度优于95%）。与参考变化结果对比，可明显看出，训练样本优化后，DBN的检测准确性显著提高，大量伪变化区域被剔除，变化与未变化区域能够被正确识别。

图 5 变化检测结果对比图(实验1)

Figure 5. Comparison of Change Detection Results (Experiment 1)

下载: 全尺寸图片幻灯片

2.2 实验2

实验2数据选自在武汉多时相场景变化检测数据集（multi-temporal scene Wuhan，MtS-WH），影像由IKONOS传感器分别获取于2002-02-11和2009-06-24，分辨率为1 m，包含红、绿、蓝和近红外4个波段，实验区影像大小为1 ‍778×1 784像素。实验区影像及RCVA、GLCM变化强度图如图 6所示，选定的RCVA和GLCM的阈值分别为79和28。优化前后变化样本标记图如图 7所示，其中红色实线框内为建筑物阴影被大量剔除的典型区域，红色虚线椭圆形内为植被季节性返青大量剔除的典型区域。

图 6 IKONOS遥感影像与变化强度图

Figure 6. IKONOS Images and Intensity Change Images

下载: 全尺寸图片幻灯片

图 7 优化前后变化样本标记图(实验2)

Figure 7. Change Sample Marker Images Before and After Executing the Optimization Strategy (Experiment 2)

下载: 全尺寸图片幻灯片

参考变化检测结果是通过MtS-WH中给出的类别标签作差，附加人工目视解译得到，类别标签作差过程完全按照MtS-WH的类别参考结果进行，可认为无相对误差，参考变化及变化检测结果如图 8所示。不同数量训练样本得到的DBN变化检测的精度如表 2所示。

图 8 变化检测结果对比图(实验2)

Figure 8. Comparison of Change Detection Results (Experiment 2)

下载: 全尺寸图片幻灯片

表 2 训练样本优化前后精度分析表(实验2)

Table 2. Precision Analysis Before and After Optimizing Training Samples (Experiment 2)

数量/个		准确率/%			召回率/%			虚警率/%			漏检率/%
正样本	负样本	A	B	C	A	B	C	A	B	C	A	B	C
5 000	5 000	73.27	83.57	86.29	64.58	73.37	77.79	45.42	27.75	23.23	35.42	26.63	22.21
10 000	10 000	72.20	83.83	85.70	66.89	77.86	79.02	47.13	28.99	25.26	33.11	22.14	20.98
30 000	30 000	72.28	84.37	86.97	57.26	79.39	81.13	46.57	28.42	23.27	42.74	20.61	18.87
50 000	50 000	73.83	83.93	86.52	65.20	80.86	82.28	44.60	29.88	24.87	34.80	19.14	17.72
80 000	80 000	72.95	84.51	87.46	63.06	81.81	84.87	45.80	29.04	23.83	36.94	18.19	15.13

下载: 导出CSV

| 显示表格

由表 2可知，实验2准确率和召回率都有较大幅度的提高，其中，准确率最大增幅为12.14%，召回率最大增幅为22.13%；同时，虚警率和漏检率都有较大幅度的下降，虚警率最大降幅为18.15%，漏检率最大降幅为22.13%。变化检测结果优化后，准确率最大增幅为2.95%，召回率最大增幅为4.42%，虚警率最大降幅为5.12%，漏检率最大降幅为4.42%。

3 结语

本文提出一种利用领域知识优化策略进行高分辨率遥感影像深度学习变化检测的方法。根据RCVA和GLCM提取的光谱和纹理特征标定初选样本；利用领域知识优化策略对样本进行筛选，获得优质样本；训练DBN模型，得到变化检测结果，并利用优化策略对检测结果进行优化。实验表明，本文方法能够有效提高变化检测结果的准确率和召回率，大幅度降低虚警率和漏检率。通过提高训练样本质量，进而提高深度学习模型检测性能，为高分辨率遥感影像深度学习变化检测提供了一条新途径。同时，在变化检测结果的基础上执行优化策略，则进一步提高了检测结果的精度。

图 1 2011—2020年高分可见光遥感图像舰船目标检测相关主题文献发表统计

Figure 1. Statistics of the Number of Publications on the Subject of Ship Detection Based on High-Resolution Optical Remote Sensing Images from 2011 to 2020

下载: 全尺寸图片幻灯片

图 2 高分可见光遥感图像舰船目标检测方法体系

Figure 2. Methodology System of Ship Detection in High-Resolution Optical Remote Sensing Images

下载: 全尺寸图片幻灯片

图 3 高分辨率可见光遥感图像舰船目标检测方法发展历程

Figure 3. Development History of Ship Detection Methods in High-Resolution Optical Remote Sensing Images

下载: 全尺寸图片幻灯片

表 1 高分辨率可见光遥感图像舰船目标检测有关公开数据集统计

Table 1 Public Detection Datasets Related to Ship Targets in High-Resolution Optical Remote Sensing

数据集	目标个数（训练+验证）	图像大小/像素	分辨率/m	标注方法	被引用次数
NWPU VHR-10^[93]	约300	450~1 500	0.5~2	水平框	261
HRSC2016 ^[94]	约3 000	300~1 500	0.4~2	水平框/旋转框	39
Kaggle-ship^[95]	约80 000	768	约2	掩码	43
HRRSD^[96]	约2 000	150~15 000	0.15~1.2	水平框	20
DOTA ^[97]	约45 000	400~15 000	0.1~5	水平框/旋转框	315
iSAID^[98]	约80 000	400~15 000	0.1~5	掩码	11
DIOR^[99]	约60 000	800	0.5~5	水平框	43
xView^[100]	约10 000	2 500~5 000	0.3	水平框	10

下载: 导出CSV

参考文献(100)

[1]	Kanjir U, Greidanus H, Oštir K. Vessel Detection and Classification from Spaceborne Optical Images: A Literature Survey[J]. Remote Sensing of Environment, 2018, 207: 1-26 doi: 10.1016/j.rse.2017.12.033
[2]	刘俊琦, 李智, 张学阳. 可见光遥感图像海面目标检测技术综述[J]. 计算机科学, 2020, 47(3): 116-123 https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA202003021.htm Liu Junqi, Li Zhi, Zhang Xueyang. Review of Maritime Target Detection in Visible Bands of Optical Remote Sensing Images[J]. Computer Science, 2020, 47(3): 116-123 https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA202003021.htm
[3]	李晓斌, 江碧涛, 杨渊博, 等. 光学遥感图像目标检测技术综述[J]. 航天返回与遥感, 2019, 40(4): 95-104 doi: 10.3969/j.issn.1009-8518.2019.04.011 Li Xiaobin, Jiang Bitao, Yang Yuanbo, et al. A Survey on Object Detection Technology in Optical Remote Sensing Images[J]. Spacecraft Recovery & Remote Sensing, 2019, 40(4): 95-104 doi: 10.3969/j.issn.1009-8518.2019.04.011
[4]	李健伟, 曲长文, 彭书娟. SAR图像舰船目标联合检测与方向估计[J]. 武汉大学学报·信息科学版, 2019, 44(6): 901-907 doi: 10.13203/j.whugis20170328 Li Jianwei, Qu Changwen, Peng Shujuan. A Joint SAR Ship Detection and Azimuth Estimation Method [J]. Geomatics and Information Science of Wuhan University, 2019, 44(6): 901-907 doi: 10.13203/j.whugis20170328
[5]	张晓东, 张力飞, 陈关州, 等. 基于深度学习的遥感影像地物目标检测和轮廓提取一体化模型[J]. 测绘地理信息, 2019, 44(6): 1-5 https://www.cnki.com.cn/Article/CJFDTOTAL-CHXG201906001.htm Zhang Xiaodong, Zhang Lifei, Chen Guanzhou, et al. An Integrated Model of Object Detection and Contour Extraction Based on Deep Learning[J]. Journal of Geomatics, 2019, 44(6): 1-5 https://www.cnki.com.cn/Article/CJFDTOTAL-CHXG201906001.htm
[6]	Kang M, Ji K F, Leng X G, et al. Contextual Region-Based Convolutional Neural Network with Multilayer Fusion for SAR Ship Detection[J]. Remote Sensing, 2017, 9(8): 860 doi: 10.3390/rs9080860
[7]	王鹏, 吕高杰, 龚俊斌, 等. 一种复杂海天背景下的红外舰船目标自动检测方法[J]. 武汉大学学报·信息科学版, 2011, 36(12): 1 438-1 441 http://ch.whu.edu.cn/article/id/743 Wang Peng, Lü Gaojie, Gong Junbin, et al. An Automatic Target Detection Method for Infrared Ship in Complex Sea-Sky Background[J]. Geomatics and Information Science of Wuhan University, 2011, 36 (12): 1 438-1 441 http://ch.whu.edu.cn/article/id/743
[8]	张志新, 徐清俊, 张川, 等. 基于单景高分四号卫星多光谱影像的舰船运动特征检测[J]. 遥感技术与应用, 2019, 34(4): 892-900 https://www.cnki.com.cn/Article/CJFDTOTAL-YGJS201904025.htm Zhang Zhixin, Xu Qingjun, Zhang Chuan, et al. Ship Moving Feature Detection Using a Single GF-4 Multispectral Image[J]. Remote Sensing Technology and Application, 2019, 34(4): 892-900 https://www.cnki.com.cn/Article/CJFDTOTAL-YGJS201904025.htm
[9]	Sun Hao, Wang Cheng, Wang Boliang. Unsupervised Video-Based Lane Detection Using LocationEnhanced Topic Models[J]. Optical Engineering, 2010, 49(10): 1-9 doi: 10.1117/1.OE.49.10.105001
[10]	LeCun Y, Bengio Y, Hinton G. Deep Learning [J]. Nature, 2015, 521(7 553): 436-444
[11]	Ren S, He K, Girshick R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(6): 1 137-1 149 http://www.ee.bgu.ac.il/~rrtammy/DNN/reading/FastSun.pdf
[12]	Redmon J, Farhadi A. YOLOv3: An Incremental Improvement[J]. ArXiv Preprint, 2018: 1804. 02767 http://users.soe.ucsc.edu/~czczycz/src/YOLOv3.pdf
[13]	Liu W, Anguelov D, Erhan D, et al. SSD: Single Shot MultiBox Detector[M]. Cham: Springer International Publishing, 2016
[14]	Lin T Y, Goyal P, Girshick R, et al. Focal Loss for Dense Object Detection[C]// IEEE International Conference on Computer Vision, Venice, Italy, 2017
[15]	Máttyus G. Near Real-Time Automatic Marine Vessel Detection on Optical Satellite Images[J]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2013, XL-1/W1: 233-237 http://elib-v3.dlr.de/82654/1/isprsarchives-XL-1-W1-233-2013.pdf
[16]	Bi F K,Liu F,Gao L N. A Hierarchical SalientRegion Based Algorithm for Ship Detection in Remote Sensing Images[J]. Advances in Neural Network Research and Applications,2010,DOI: 10.1007/978-3-642-12990-2_85
[17]	Liu G, Zhang Y S, Zheng X W, et al. A New Method on Inshore Ship Detection in High-Resolution Satellite Images Using Shape and Context Information[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(3): 617-621 doi: 10.1109/LGRS.2013.2272492
[18]	Proia N, Pagé V. Characterization of a Bayesian Ship Detection Method in Optical Satellite Images [J]. IEEE Geoscience and Remote Sensing Letters, 2010, 7(2): 226-230 doi: 10.1109/LGRS.2009.2031826
[19]	Zhu C R, Zhou H, Wang R S, et al. A Novel Hierarchical Method of Ship Detection from Spaceborne Optical Image Based on Shape and Texture Features[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(9): 3 446-3 456 doi: 10.1109/TGRS.2010.2046330
[20]	祝胜男, 郭炜炜, 柳彬, 等. 利用超像素级上下文特征进行靠岸集装箱船检测[J]. 武汉大学学报·信息科学版, 2019, 44(4): 578-585 doi: 10.13203/j.whugis20170129 Zhu Shengnan, Guo Weiwei, Liu Bin, et al. Detection Containers Based on Superpixel-Level Contextual Feature[J]. Geomatics and Information Science of Wuhan University, 2019, 44(4): 578-585 doi: 10.13203/j.whugis20170129
[21]	Wang B Y, Zhang R, Yuan Y, et al. An AutoAdapt Multi-level Threshold Segmentation Method of Ships Detection in Remote Sensing Images with Complex Sea Surface Background[C]//2011 International Workshop on Multi-Platform/Multi-Sensor Remote Sensing and Mapping, Xiamen, China, 2011
[22]	Kumar S, Selvi M. Sea Object Detection Using Colour and Texture Classification[J]. International Journal of Computer Applications in Technology, 2011, 1(1): 59-63
[23]	Rainey K, Stastny J. Object Recognition in Ocean Imagery Using Feature Selection and Compressive Sensing[C]// IEEE Applied Imagery Pattern Recognition Workshop (AIPR), Washington DC, America, 2011
[24]	Bi F K, Zhu B C, Gao L N, et al. A Visual Search Inspired Computational Model for Ship Detection in Optical Satellite Images[J]. IEEE Geoscience and Remote Sensing Letters, 2012, 9(4): 749-753 doi: 10.1109/LGRS.2011.2180695
[25]	Li Z C, Itti L. Saliency and Gist Features for Target Detection in Satellite Images[J]. IEEE Transactions on Image Processing, 2011, 20(7): 2 017- 2 029 doi: 10.1109/TIP.2010.2099128
[26]	Han J W, He S, Qian X L, et al. An ObjectOriented Visual Saliency Detection Framework Based on Sparse Coding Representations[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2013, 23(12): 2 009-2 021 doi: 10.1109/TCSVT.2013.2242594
[27]	Zhao Q, Koch C. Learning Saliency-Based Visual Attention: A Review[J]. Signal Processing, 2013, 93(6): 1 401-1 407 doi: 10.1016/j.sigpro.2012.06.014
[28]	Xia Y, Wan S H, Jin P Q, et al. A Novel SeaLand Segmentation Algorithm Based on Local Binary Patterns for Ship Detection[J]. International Journal of Signal Processing, Image Processing and Pattern Recognition, 2014, 7(3): 237-246 doi: 10.14257/ijsip.2014.7.3.19
[29]	Li S, Zhou Z Q, Wang B, et al. A Novel Inshore Ship Detection via Ship Head Classification and Body Boundary Determination[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(12): 1 920- 1 924 doi: 10.1109/LGRS.2016.2618385
[30]	Ma L, Guo J, Wang Y Q, et al. Ship Detection by Salient Convex Boundaries[C]//The 3rd International Congress on Image and Signal Processing, Yantai, China, 2010
[31]	Borji A, Cheng M M, Hou Q B, et al. Salient Object Detection: A Survey[J]. Computational Visual Media, 2019, 5(2): 117-150 doi: 10.1007/s41095-019-0149-9
[32]	Han J W, Zhou P C, Zhang D W, et al. Efficient, Simultaneous Detection of Multi-class Geospatial Targets Based on Visual Saliency Modeling and Discriminative Learning of Sparse Coding[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 89: 37-48 doi: 10.1016/j.isprsjprs.2013.12.011
[33]	郭少军, 沈同圣, 徐健, 等. 基于ObjectNess BING的海面多舰船目标检测[J]. 系统工程与电子技术, 2016, 38(1): 14-20 https://www.cnki.com.cn/Article/CJFDTOTAL-XTYD201601004.htm Guo Shaojun, Shen Tongsheng, Xu Jian, et al. Detection of Multi-Ship Targets at Sea Based on Object Ness BING[J]. Systems Engineering and Electronics, 2016, 38(1): 14-20 https://www.cnki.com.cn/Article/CJFDTOTAL-XTYD201601004.htm
[34]	Jubelin G, Khenchaf A. Multiscale Algorithm for Ship Detection in Mid, High and very High Resolution Optical Imagery[C]// IEEE Geoscience and Remote Sensing Symposium, Quebec City, Canada, 2014
[35]	Yang G, Li B, Ji S F, et al. Ship Detection from Optical Satellite Images Based on Sea Surface Analysis[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(3): 641-645 doi: 10.1109/LGRS.2013.2273552
[36]	Li Z M, Yang D Q, Chen Z Z. Multi-Layer Sparse Coding Based Ship Detection for Remote Sensing Images[C]// IEEE International Conference on Information Reuse and Integration, San Francisco, California, USA, 2015
[37]	Qi S X, Ma J, Lin J, et al. Unsupervised Ship Detection Based on Saliency and S-HOG Descriptor from Optical Satellite Images[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(7): 1 451- 1 455 doi: 10.1109/LGRS.2015.2408355
[38]	Yang F, Xu Q Z, Gao F, et al. Ship Detection from Optical Satellite Images Based on Visual Search Mechanism[C]// IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Milan, Italy, 2015
[39]	Sui H G,Song Z N. A Novel Ship Detection Method for Large-Scale Optical Satellite Images Based on Visual LBP Feature and Visual Attention Model [J]. ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2016,DOI: 10.5194/isprsarchives-XLI-B3-917-2016
[40]	Wang X, Shen S, Ning C, et al. Multi-Class Remote Sensing Object Recognition Based on Discriminative Sparse Representation[J]. Applied Optics, 2016, 55(6): 1 381-1 394 doi: 10.1364/AO.55.001381
[41]	Xu F, Liu J H. Ship Detection and Extraction Using Visual Saliency and Histogram of Oriented Gradient [J]. Optoelectronics Letters, 2016, 12(6): 473-477 doi: 10.1007/s11801-016-6179-y
[42]	Xu F, Liu J H, Dong C, et al. Ship Detection in Optical Remote Sensing Images Based on Wavelet Transform and Multi-Level False Alarm Identification[J]. Remote Sensing, 2017, 9(10): 985 doi: 10.3390/rs9100985
[43]	Nie T, He B, Bi G L, et al. A Method of Ship Detection Under Complex Background[J]. ISPRS International Journal of Geo Information, 2017, 6 (6): 159 doi: 10.3390/ijgi6060159
[44]	Zou Z X, Shi Z W. Ship Detection in Spaceborne Optical Image with SVD Networks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(10): 5 832-5 845 doi: 10.1109/TGRS.2016.2572736
[45]	Yokoya N, Iwasaki A. Object Detection Based on Sparse Representation and Hough Voting for Optical Remote Sensing Imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(5): 2 053-2 062 doi: 10.1109/JSTARS.2015.2404578
[46]	He H J, Lin Y D, Chen F, et al. Inshore Ship Detection in Remote Sensing Images via Weighted Pose Voting[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(6): 3 091-3 107 doi: 10.1109/TGRS.2017.2658950
[47]	Song P F, Qi L, Qian X M, et al. Detection of Ships in Inland River Using High-Resolution Optical Satellite Imagery Based on Mixture of Deformable Part Models[J]. Journal of Parallel and Distributed Computing, 2019, 132: 1-7 doi: 10.1016/j.jpdc.2019.04.013
[48]	Alom M Z, Taha T M, Yakopcic C, et al. The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches[J]. ArXiv Preprint , 2018: 1803. 01164 http://arxiv.org/ftp/arxiv/papers/1803/1803.01164.pdf
[49]	张瑞倩, 邵振峰, Aleksei Portnov, 等. 多尺度空洞卷积的无人机影像目标检测方法[J]. 武汉大学学报·信息科学版, 2020, 45(6): 895-903 doi: 10.13203/j.whugis20200253 Zhang Ruiqian, Shao Zhenfeng, Portnov A, et al. Multi-scale Dilated Convolutional Neural Network for Object Detection in UAV Images[J]. Geomatics and Information Science of Wuhan University, 2020, 45(6): 895-903 doi: 10.13203/j.whugis20200253
[50]	Yu F, Koltun V. Multi-scale Context Aggregation by Dilated Convolutions[J]. ArXiv Preprint, 2015: 1511. 07122 http://www.cvlibs.net/projects/autonomous_vision_survey/slides/Yu2016ICLR/top.pdf
[51]	Chen L C, Papandreou G, Kokkinos I, et al. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFS[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40 (4): 834-848 doi: 10.1109/TPAMI.2017.2699184
[52]	Zheng Z, Zhong Y F, Ma A L, et al. HyNet: Hyper-Scale Object Detection Network Framework for Multiple Spatial Resolution Remote Sensing Imagery [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 166: 1-14 doi: 10.1016/j.isprsjprs.2020.04.019
[53]	Xie W Y, Qin H N, Li Y S, et al. A Novel Effectively Optimized One-Stage Network for Object Detection in Remote Sensing Imagery[J]. Remote Sensing, 2019, 11(11): 1 376 doi: 10.3390/rs11111376
[54]	Chen C Y, Gong W G, Chen Y L, et al. Object Detection in Remote Sensing Images Based on a SceneContextual Feature Pyramid Network[J]. Remote Sensing, 2019, 11(3): 339 doi: 10.3390/rs11030339
[55]	Hu J, Shen L, Sun G. Squeeze-and-Excitation Networks[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, Utah, USA, 2018
[56]	Jaderberg M, Simonyan K, Zisserman A. Spatial Transformer Networks[C]// Advances in Neural Information Processing Systems, Montreal, Canada, 2015
[57]	Woo S, Park J, Lee J Y, et al. CBAM: Convolutional Block Attention Module[M]. Cham: Springer International Publishing, 2018
[58]	Wang X L, Girshick R, Gupta A, et al. Non-Local Neural Networks[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, Utah, USA, 2018
[59]	Huang G Q, Wan Z N, Liu X G, et al. Ship Detection Based on Squeeze Excitation Skip-Connection Path Networks for Optical Remote Sensing Images [J]. Neurocomputing, 2019, 332: 215-223 doi: 10.1016/j.neucom.2018.12.050
[60]	Al-Alimi D, Shao Y X, Feng R Y, et al. MultiScale Geospatial Object Detection Based on ShallowDeep Feature Extraction[J]. Remote Sensing, 2019, 11(21): 2 525 doi: 10.3390/rs11212525
[61]	Azimi S M, Vig E, Bahmanyar R, et al. Towards Multi-Class Object Detection in Unconstrained Remote Sensing Imagery[C]. The 14th Asian Conference on Computer Vision, Perth, Australia, 2018
[62]	Yang X, Yang J R, Yan J C, et al. SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated Objects[C]// IEEE/CVF International Conference on Computer Vision, Seoul, Korea (South), 2019
[63]	Zhang G J, Lu S J, Zhang W. CAD-Net: A Context-Aware Detection Network for Objects in Remote Sensing Imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(12): 10 015-10 024 doi: 10.1109/TGRS.2019.2930982
[64]	Wang J W, Ding J, Guo H W, et al. Mask OBB: A Semantic Attention-Based Mask Oriented Bounding Box Representation for Multi-category Object Detection in Aerial Images[J]. Remote Sensing, 2019, 11 (24): 2 930 doi: 10.3390/rs11242930
[65]	Li C, Xu C, Cui Z, et al. Learning Object-Wise Semantic Representation for Detection in Remote Sensing Imagery[C]. The IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, California, USA, 2019
[66]	Lin H N, Shi Z W, Zou Z X. Fully Convolutional Network with Task Partitioning for Inshore Ship Detection in Optical Remote Sensing Images[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (10): 1 665-1 669 doi: 10.1109/LGRS.2017.2727515
[67]	Lin T Y, Dollár P, Girshick R, et al. Feature Pyramid Networks for Object Detection[C]// IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, Hawaii, USA, 2017
[68]	Li Q P, Mou L C, Liu Q J, et al. HSF-Net: Multiscale Deep Feature Embedding for Ship Detection in Optical Remote Sensing Imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56 (12): 7 147-7 161 doi: 10.1109/TGRS.2018.2848901
[69]	Yang X, Sun H, Fu K, et al. Automatic Ship Detection in Remote Sensing Images from Google Earth of Complex Scenes Based on Multiscale Rotation Dense Feature Pyramid Networks[J]. Remote Sensing, 2018, 10(1): 132 doi: 10.3390/rs10010132
[70]	Deng Z P, Sun H, Zhou S L, et al. Multi-Scale Object Detection in Remote Sensing Imagery with Convolutional Neural Networks[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 145: 3-22 doi: 10.1016/j.isprsjprs.2018.04.003
[71]	Song Z N, Sui H G, Hua L. A Hierarchical Object Detection Method in Large-Scale Optical Remote Sensing Satellite Imagery Using Saliency Detection and CNN[J]. International Journal of Remote Sensing, 2021, 42(8): 2 827-2 847 doi: 10.1080/01431161.2020.1826059
[72]	Liu S, Qi L, Qin H F, et al. Path Aggregation Network for Instance Segmentation[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, Utah, USA, 2018
[73]	Ghiasi G, Lin T Y, Le Q V. NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, California, USA, 2019
[74]	徐志京, 丁莹. 自适应旋转区域生成网络的遥感图像舰船目标检测[J]. 激光与光电子学进展, 2020, 57(24): 408-415 https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202024046.htm Xu Zhijing, Ding Ying. Ship Object Detection of Remote Sensing Images Based on Adaptive Rotation Region Proposal Network[J]. Laser & Optoelectronics Progress, 2020, 57(24): 408-415 https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ202024046.htm
[75]	Liu L, Pan Z X, Lei B. Learning a Rotation Invariant Detector with Rotatable Bounding Box[J]. ArXiv Preprint, 2017: 1711. 09405 http://arxiv.org/pdf/1711.09405
[76]	Liu Z K, Hu J G, Weng L B, et al. Rotated Region Based CNN for Ship Detection[C]// IEEE International Conference on Image Processing(ICIP), Beijing, China, 2017
[77]	Zhang Z H, Guo W W, Zhu S N, et al. Toward Arbitrary-Oriented Ship Detection with Rotated Region Proposal and Discrimination Networks[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(11): 1 745-1 749 doi: 10.1109/LGRS.2018.2856921
[78]	Yang X, Liu Q Q, Yan J C, et al. R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object[J]. ArXiv Preprint , 2019: 1908. 05612 http://arxiv.org/abs/1908.05612v1
[79]	Ding J, Xue N, Long Y, et al. Learning RoI Transformer for Oriented Object Detection in Aerial Images [C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, California, USA, 2019
[80]	Zhu Y X, Du J, Wu X Q. Adaptive Period Embedding for Representing Oriented Objects in Aerial Images [J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(10): 7 247-7 257 doi: 10.1109/TGRS.2020.2981203
[81]	Dong Z P, Wang M, Wang Y L, et al. Object Detection in High Resolution Remote Sensing Imagery Based on Convolutional Neural Networks with Suitable Object Scale Features[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58 (3): 2 104-2 114 doi: 10.1109/TGRS.2019.2953119
[82]	ZhangY J, ShengW G, Jiang JF, et al. PriorityBranches for Ship Detection in Optical Remote Sensing Images [J]. Remote Sensing, 2020, 12(7): 1 196 doi: 10.3390/rs12071196
[83]	Yu Y T, Guan H Y, Li D L, et al. Orientation Guided Anchoring for Geospatial Object Detection from Remote Sensing Imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 160: 67-82 doi: 10.1016/j.isprsjprs.2019.12.001
[84]	Sun S H, Lu Z X, Liu W J, et al. Shipnet for Semantic Segmentation on VHR Maritime Imagery [C]// IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 2018
[85]	Bao S Z, Zhong X, Zhu R F, et al. Single Shot Anchor Refinement Network for Oriented Object Detection in Optical Remote Sensing Imagery[J]. IEEE Access, 2019, 7: 87 150-87 161 doi: 10.1109/ACCESS.2019.2924643
[86]	Dong R C, Xu D Z, Zhao J, et al. Sig-NMSBased Faster R-CNN Combining Transfer Learning for Small Target Detection in VHR Optical Remote Sensing Imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(11): 8 534-8 545 doi: 10.1109/TGRS.2019.2921396
[87]	Li Z Z, You Y N, Liu F. Multi-Scale Ships Detection in High-Resolution Remote Sensing Image via Saliency-Based Region Convolutional Neural Network[C]// International Geoscience and Remote Sensing Symposium, Yokohama, Japan, 2019
[88]	You Y N, Li Z Z, Ran B H, et al. Broad Area Target Search System for Ship Detection via Deep Convolutional Neural Network[J]. Remote Sensing, 2019, 11(17): 1 965 doi: 10.3390/rs11171965
[89]	Yu Y, Ai H, He X J, et al. Ship Detection in Optical Satellite Images Using Haar-Like Features and Periphery-Cropped Neural Networks[J]. IEEE Access, 2018, 6: 71 122-71 131 doi: 10.1109/ACCESS.2018.2881479
[90]	Wang Y L, Dong Z P, Zhu Y. Multiscale Block Fusion Object Detection Method for Large-Scale HighResolution Remote Sensing Imagery[J]. IEEE Access, 2019, 7: 99 530-99 539 doi: 10.1109/ACCESS.2019.2930092
[91]	He T, Zhang Z, Zhang H, et al. Bag of Tricks for Image Classification with Convolutional Neural Networks[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 2019
[92]	Zhang Z, Cui X, Zheng Q, et al. Land Use Classification of Remote Sensing Images Based on Convolution Neural Network[J]. Arabian Journal of Geosciences, 2021, 14(4): 1-6 doi: 10.1007/s12517-021-06587-5
[93]	Cheng G, Han J W, Zhou P C, et al. Multi-Class Geospatial Object Detection and Geographic Image Classification Based on Collection of Part Detectors [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 98: 119-132 doi: 10.1016/j.isprsjprs.2014.10.002
[94]	Liu Z K, Yuan L, Weng L B, et al. A High Resolution Optical Satellite Image Dataset for Ship Recognition and Some New Baselines[C]// The 6th International Conference on Pattern Recognition Applications and Methods, Porto, Portugal, 2017
[95]	Polat M, Alameri H, Oral E A, et al. Ship Detection in Satellite Images[C]// ISASE 2018, Spokane, Washington DC, USA, 2018
[96]	Zhang Y L, Yuan Y, Feng Y C, et al. Hierarchical and Robust Convolutional Neural Network for very High-Resolution Remote Sensing Object Detection [J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(8): 5 535-5 548 doi: 10.1109/TGRS.2019.2900302
[97]	Xia G S, Bai X, Ding J, et al. DOTA: A LargeScale Dataset for Object Detection in Aerial Images [C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, Utah, USA, 2018
[98]	Zamir S W, Arora A, Gupta A, et al. iSAID: A Large-Scale Dataset for Instance Segmentation in Aerial Images[C]//IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, California, USA, 2019
[99]	Li K, Wan G, Cheng G, et al. Object Detection in Optical Remote Sensing Images: A Survey and a New Benchmark[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 159: 296-307 doi: 10.1016/j.isprsjprs.2019.11.023
[100]	Lam D, Kuzma R, McGee K, et al. xView: Objects in Context in Overhead Imagery[J]. ArXiv Preprint, 2018: 1802. 07856 http://arxiv.org/pdf/1802.07856

施引文献(11)

期刊类型引用(7)

1.	刘立，董先敏，王德富，张志强，刘娟. 人机协同的多模态遥感变化检测方法. 测绘通报. 2024(S1): 130-136 . 百度学术
2.	刘莺迎，周刚. 基于孪生Transformer的双时相遥感影像变化检测方法. 遥感信息. 2024(03): 67-74 . 百度学术
3.	张永军，王飞，李彦胜，欧阳松，魏东，刘晓建，孔德宇，CHEN Ruixian，ZHANG Bin. 遥感知识图谱创建及其典型场景应用技术. 遥感学报. 2023(02): 249-266 . 百度学术
4.	周豫阳，王明常，王凤艳，杨盈，刘子维. 改进U-Net的高分辨率遥感影像建筑区变化检测方法. 世界地质. 2023(01): 159-167 . 百度学术
5.	陈桥驿，闫宇飞，黄永芳. 融合面向对象和深度置信网络的农村建筑物信息提取. 测绘通报. 2023(06): 134-137 . 百度学术
6.	张征稳. 面向建筑物边缘特征保持的遥感影像变化检测研究. 测绘与空间地理信息. 2023(S1): 203-207 . 百度学术
7.	梁宇，左栋. “问题地图”智能检测发展现状与关键技术. 测绘通报. 2023(10): 111-116 . 百度学术

其他类型引用(4)

资源附件(0)

图(3) / 表(1)

计量

文章访问数:
HTML全文浏览量:
PDF下载量:
被引次数: 11

1 本文方法
1.1 光谱变化特征提取
1.2 纹理变化特征提取
1.3 结合领域知识的训练样本优化
1.4 DBN模型训练与变化检测
2 实验与分析
2.1 实验1
2.2 实验2
3 结语

高分辨率可见光遥感图像舰船目标检测综述

作者简介: 宋志娜，博士，主要从事遥感图像目标信息提取研究。song.zhina@163.com

通讯作者: 眭海刚，博士，教授。haigang_sui@263.net

计量

出版历程